Refine prompts and generate images
Generate tags for images using Waifu Diffusion models
The beginning of large-scale illustration-sdxl model